skactiveml.classifier.ParzenWindowClassifier#

class skactiveml.classifier.ParzenWindowClassifier(n_neighbors=None, metric='rbf', metric_dict=None, classes=None, missing_label=nan, cost_matrix=None, class_prior=0.0, random_state=None)[source]#

Bases: ClassFrequencyEstimator

The Parzen window classifier (PWC) is a simple and probabilistic classifier. This classifier is based on a non-parametric density estimation obtained by applying a kernel function.

Parameters

classesarray-like of shape (n_classes), default=None: Holds the label for each class. If none, the classes are determined during the fit.
missing_labelscalar or string or np.nan or None, default=np.nan: Value to represent a missing label.
cost_matrixarray-like of shape (n_classes, n_classes), default=None: Cost matrix with cost_matrix[i,j] indicating cost of predicting class classes[j] for a sample of class classes[i]. Can be only set, if classes is not none.
class_priorfloat or array-like of shape (n_classes,), default=0: Prior observations of the class frequency estimates. If class_prior is an array, the entry class_prior[i] indicates the non-negative prior number of samples belonging to class classes_[i]. If class_prior is a float, class_prior indicates the non-negative prior number of samples per class.
metricstr or callable, default=’rbf’: The metric must be a valid kernel defined by the function sklearn.metrics.pairwise.pairwise_kernels.
n_neighborsint or None, default=None: Number of nearest neighbours. Default is None, which means all available samples are considered.
metric_dictdict, default=None: Any further parameters are passed directly to the kernel function. For the kernel ‘rbf’ we allow the use of mean kernel [2] and use it when gamma is set to ‘mean’ (i.e., {‘gamma’: ‘mean’}). While N is defined as the labeled data the variance is calculated over all X.
random_stateint or RandomState instance or None, default=None: Determines random number for ‘predict’ method. Pass an int for reproducible results across multiple method calls.

References

1: O. Chapelle, “Active Learning for Parzen Window Classifier”, Proceedings of the Tenth International Workshop Artificial Intelligence and Statistics, 2005.
2: Chaudhuri, A., Kakde, D., Sadek, C., Gonzalez, L., & Kong, S., “The Mean and Median Criteria for Kernel Bandwidth Selection for Support Vector Data Description” IEEE International Conference on Data Mining Workshops (ICDMW), 2017.

Attributes

classes_array-like of shape (n_classes,): Holds the label for each class after fitting.
class_priornp.ndarray of shape (n_classes): Prior observations of the class frequency estimates. The entry class_prior_[i] indicates the non-negative prior number of samples belonging to class classes_[i].
cost_matrix_np.ndarray of shape (classes, classes): Cost matrix with cost_matrix_[i,j] indicating cost of predicting class classes_[j] for a sample of class classes_[i].
X_np.ndarray of shape (n_samples, n_features): The sample matrix X is the feature matrix representing the samples.
V_np.ndarray of shape (n_samples, classes): The class labels are represented by counting vectors. An entry V[i,j] indicates how many class labels of classes[j] were provided for training sample X_[i].

Methods

`fit`(X, y[, sample_weight])	Fit the model using X as training data and y as class labels.
`get_metadata_routing`()	Get metadata routing of this object.
`get_params`([deep])	Get parameters for this estimator.
`predict`(X)	Return class label predictions for the test samples X.
`predict_freq`(X)	Return class frequency estimates for the input samples 'X'.
`predict_proba`(X)	Return probability estimates for the test data X.
`sample_proba`(X[, n_samples, random_state])	Samples probability vectors from Dirichlet distributions whose parameters alphas are defined as the sum of the frequency estimates returned by predict_freq and the class_prior.
`score`(X, y[, sample_weight])	Return the mean accuracy on the given test data and labels.
`set_fit_request`(*[, sample_weight])	Request metadata passed to the `fit` method.
`set_params`(**params)	Set the parameters of this estimator.
`set_score_request`(*[, sample_weight])	Request metadata passed to the `score` method.

Attributes

METRICS

METRICS = ['additive_chi2', 'chi2', 'cosine', 'linear', 'poly', 'polynomial', 'rbf', 'laplacian', 'sigmoid', 'precomputed']#

fit(X, y, sample_weight=None)[source]#

Fit the model using X as training data and y as class labels.

Parameters

Xarray-like of shape (n_samples, n_features): The sample matrix X is the feature matrix representing the samples.
yarray-like of shape (n_samples): It contains the class labels of the training samples.
sample_weightarray-like of shape (n_samples), default=None: It contains the weights of the training samples’ class labels. It must have the same shape as y.

Returns

self: ParzenWindowClassifier,: The ParzenWindowClassifier is fitted on the training data.

get_metadata_routing()#

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns

routingMetadataRequest: A MetadataRequest encapsulating routing information.

get_params(deep=True)#

Get parameters for this estimator.

Parameters

deepbool, default=True: If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

paramsdict: Parameter names mapped to their values.

predict(X)#

Return class label predictions for the test samples X.

Parameters

Xarray-like of shape (n_samples, n_features): Input samples.

Returns

ynumpy.ndarray of shape (n_samples): Predicted class labels of the test samples X. Classes are ordered according to classes_.

predict_freq(X)[source]#

Return class frequency estimates for the input samples ‘X’.

Parameters

X: array-like or shape (n_samples, n_features) or shape (n_samples, m_samples) if metric == ‘precomputed’: Input samples.

Returns

F: array-like of shape (n_samples, classes): The class frequency estimates of the input samples. Classes are ordered according to classes_.

predict_proba(X)#

Return probability estimates for the test data X.

Parameters

Xarray-like, shape (n_samples, n_features) or
shape (n_samples, m_samples) if metric == ‘precomputed’: Input samples.

Returns

Parray-like of shape (n_samples, classes): The class probabilities of the test samples. Classes are ordered according to classes_.

sample_proba(X, n_samples=10, random_state=None)#

Samples probability vectors from Dirichlet distributions whose parameters alphas are defined as the sum of the frequency estimates returned by predict_freq and the class_prior.

Parameters

Xarray-like of shape (n_test_samples, n_features): Test samples for which n_samples probability vectors are to be sampled.
n_samplesint, default=10: Number of probability vectors to sample for each X[i].
random_stateint or numpy.random.RandomState or None, default=None: Ensure reproducibility when sampling probability vectors from the Dirichlet distributions.

Returns

Parray-like of shape (n_samples, n_test_samples, n_classes): There are n_samples class probability vectors for each test sample in X. Classes are ordered according to classes_.

score(X, y, sample_weight=None)#

Return the mean accuracy on the given test data and labels.

Parameters

Xarray-like of shape (n_samples, n_features): Test samples.
yarray-like of shape (n_samples,): True labels for X.
sample_weightarray-like of shape (n_samples,), default=None: Sample weights.

Returns

scorefloat: Mean accuracy of self.predict(X) regarding y.

set_fit_request(*, sample_weight: Union[bool, None, str] = '$UNCHANGED$') → ParzenWindowClassifier#

Request metadata passed to the fit method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.
False: metadata is not requested and the meta-estimator will not pass it to fit.
None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.
str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters

sample_weightstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for sample_weight parameter in fit.

Returns

selfobject: The updated object.

set_params(**params)#

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**paramsdict: Estimator parameters.

Returns

selfestimator instance: Estimator instance.

set_score_request(*, sample_weight: Union[bool, None, str] = '$UNCHANGED$') → ParzenWindowClassifier#

Request metadata passed to the score method.

Note that this method is only relevant if enable_metadata_routing=True (see sklearn.set_config()). Please see User Guide on how the routing mechanism works.

The options for each parameter are:

True: metadata is requested, and passed to score if provided. The request is ignored if metadata is not provided.
False: metadata is not requested and the meta-estimator will not pass it to score.
None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.
str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Note

This method is only relevant if this estimator is used as a sub-estimator of a meta-estimator, e.g. used inside a Pipeline. Otherwise it has no effect.

Parameters

sample_weightstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED: Metadata routing for sample_weight parameter in score.

Returns

selfobject: The updated object.

Examples using `skactiveml.classifier.ParzenWindowClassifier`#

Core Set

Random Sampling

Active Learning with Cost Embedding

Clustering Uncertainty-weighted Embeddings (CLUE)

Contrastive Active Learning (CAL)

Batch Active Learning by Diverse Gradient Embedding (BADGE)

Typical Clustering (TypiClust)

Uncertainty Sampling with Entropy

Epistemic Uncertainty Sampling

Querying Informative and Representative Examples

Uncertainty Sampling with Margin

Uncertainty Sampling with Least-Confidence

Probability Coverage (ProbCover)

Expected Average Precision

Monte-Carlo EER with Log-Loss

Monte-Carlo EER with Misclassification-Loss

Sub-sampling Wrapper

Discriminative Active Learning

Parallel Utility Estimation Wrapper

Multi-class Probabilistic Active Learning

Query-by-Committee (QBC) with Vote Entropy

Query-by-Committee (QBC) with Variation Ratios

Query-by-Committee (QBC) with Kullback-Leibler Divergence

Value of Information

Value of Information on Labeled Samples

Value of Information on Unlabeled Samples

Split

Periodic Sampling

Stream Random Sampling

Fixed-Uncertainty

Variable-Uncertainty

Density Based Active Learning for Data Streams

Randomized-Variable-Uncertainty

Cognitive Dual-Query Strategy with Random Sampling

Cognitive Dual-Query Strategy with Fixed-Uncertainty

Cognitive Dual-Query Strategy with Variable-Uncertainty

Cognitive Dual-Query Strategy with Randomized-Variable-Uncertainty

Probabilistic Active Learning in Datastreams

skactiveml.classifier.ParzenWindowClassifier#

Examples using skactiveml.classifier.ParzenWindowClassifier#

Examples using `skactiveml.classifier.ParzenWindowClassifier`#